Tutorial · Task guide
Fix ‘The Character Forgot’: Memory, Lorebook, Context, and Prompt Checks
Before you start
- A saved chat where the missing fact can be reproduced without sending sensitive credentials.
- Access to the card's active lorebook and long-chat summary/settings.
- A clear one-sentence statement of the fact and when it should apply.
Where to find it
Chat > ⋮ > Current status > World loreLong-press message > PromptChat > ⋮ > All tools > Long-chat summaryChat > ⋮ > Current status > Persona
Step-by-step
Classify the forgotten fact
Put identity and stable world rules in the card/lorebook; session events in history or summary; user identity in Persona; temporary scene instructions in Author's Note/system guidance.
Reproduce with a neutral question
Ask one direct, non-leading question in a copy or checkpointed session. Avoid adding the answer to the question, because that hides whether retrieval worked.
Inspect the actual prompt
Long-press the reply and tap Prompt. Search Prompt Inspector for the exact fact, the relevant lore entry name/key, summary, Persona, and any conflicting instruction. Edit active lore under chat ⋮ → Current status → World lore.
Repair the missing layer
If a lore entry is absent, enable the book, fix primary/secondary keys, scan depth, probability, recursion, position, or token budget. If old session events are absent, refresh the long-chat summary or preserve a checkpoint.
Test once, then remove duplication
Send the same neutral test after the repair. When it works, remove duplicate constant entries or contradictory summaries so context remains stable and cheaper.
How to know it worked
- Prompt Inspector contains the correct single source for the fact.
- The character answers the neutral test consistently for at least two nearby turns.
- No contradictory duplicate remains in card, lorebook, summary, and Persona.
- Context/token use stays within the chosen model's limits.
Cost, data, and irreversible-risk notes
- Editing cards, lorebooks, summaries, and checkpoints is local. Every diagnostic reply/regeneration is still a model call and may be billed.
- Making every lore entry constant increases tokens and can crowd out recent conversation, raising cost while making memory worse.
- Prompt Inspector can expose private card/history content. Redact it before sharing; API keys should never appear there or in screenshots.
Troubleshooting
The lore entry never appears in Prompt Inspector
Likely cause: Book/entry disabled, key mismatch, secondary-key rule, probability, scan depth, recursion, or budget excluded it.
What to do: Test the exact key against recent message text, temporarily make activation deterministic, verify position/budget, then restore intentional probability after the entry is proven.
The fact is present but the model still contradicts it
Likely cause: A stronger conflicting instruction exists, wording is ambiguous, or the model/context is weak for the task.
What to do: Remove contradictions, state the rule once in imperative/plain language, place it in the appropriate priority layer, and test with a capable model.
Recent facts work, old events vanish
Likely cause: The raw history fell outside the context window and the summary omitted the event.
What to do: Regenerate/edit the long-chat summary to include the durable event, create a checkpoint, or move a truly permanent fact into lore rather than inflating raw history.
Context too long / token limit
Likely cause: Too many constants, oversized card/lore, duplicate summaries, or long output reservation consumes the model window.
What to do: Trim duplicates, turn situational constants back into keyed entries, shorten summaries, and use a model with a larger context only when needed.
Frequently asked questions
Should every important fact be a constant lore entry?
No. Constants always consume context. Use them only for rules that must apply every turn; use triggers for situational lore and summaries/history for session events.
What is the fastest way to prove a worldbook worked?
Trigger it with known text, then open Prompt Inspector on the resulting reply and search for the entry content. A plausible answer alone is not proof.
Is character memory the same as Companion memory?
No. Roleplay sessions rely on card, lorebook, history and summaries. Companion adds an explicit editable long-term-memory layer; diagnose the surface you are actually using.
Sources and verification
- Roleplay Chat guide · Foreverse · Verified
- Worldbook Runtime spec · Foreverse · Verified
Coverage and verification
| Platform | Android roleplay |
|---|---|
| App version | v76+ |
| Path verified |
v55 · Diagnose a worldbook trigger with a real control
Published
Published on this site (self-hosted): Chinese edition on 2026-08-12, English edition on 2026-08-13 (English narration and captions; the app capture shows the Chinese UI). Real capture of the official Android build with waits trimmed; features and results are not staged. The narration is AI-synthesized.
- Result, prerequisites, and cost
- Prepare four lore entries and verify enabled state
- Change only one keyword
- Negative control: two constant entries
- Positive test: two entries become three
- Interpretation and troubleshooting
- Evidence checklist recap